We are living in an age when most of us interact with one computer or another every single day. With the explosion of devices and the internet of things, and the artificial intelligence that has been unleashed upon us, that interaction will only keep growing as time goes on.
As devices become smarter and come equipped with state-of-the-art technology, the way we communicate with them has also been constantly evolving. One of the earliest such changes that fundamentally altered the terms of communication between humans and computers came in the form of Fortran. The team behind Fortran, which changed programming forever, was led by American computer scientist John Backus, also referred to as the father of Fortran. Growing up, he never showed any particular talent, save perhaps a willingness to be ill-disciplined. As a student of the prestigious Hill School in Pottstown, Pennsylvania, Backus was more busy looking for the next rule to break, than anything else. Different people in the team might have been driven for different reasons, but they collectively attained their objective, step by step. The way humans communicated with computers was about to be changed forever. Data-laden problems could be handled effectively in Fortran, be it for computing economic and statistical models, missile trajectories, or flight patterns for NASA. For years, Backus and his team continued to tinker with Fortran until the compiler attained a degree of stability. Fortran first gained a foothold in the scientific community, before becoming the dominant programming language.
John Backus | Photo Credit: PIerre.Lescanne / Wikimedia Commons Born on December 3, 1924, in Philadelphia, Pennsylvania, Backus was the son of a wealthy stockbroker. On September 20, 1954, they had one of their first successes, as the first successful test compilation and execution of a computer program using Fortran took place. Later that same year, the team had enough confidence in their work to release a preliminary report on November 10. The IBM Mathematical Formula Translating System, or Fortran, took a couple of years to go from conception to release and made its debut in 1957. What had until then been a tough task of manually punching in as many as 1,000 program instructions for any particular problem could suddenly be achieved through less than 100 in Fortran. Soon, every IBM 704 was shipped with the Fortran program and its manual. Nearly seven decades on, Fortran still remains in vogue, with its most recent stable release having taken place in 2023.
Drafted into the U.S. Army, he stayed in the country for what remained of World War II. His excellent results in the military aptitude tests meant that he was first sent to an engineering programme at the University of Pittsburgh and then to Haverford College to study medicine. A Fortran punch card used by the European Space Agency. | Photo Credit: Till Niermann / Wikimedia Commons It was during this period that he was diagnosed with a cranial bone tumour that was removed and replaced with a metal plate. Without a clue as to what his future held, he decided to settle in New York City. He wished to build a hi-fi set for himself and, with that objective, entered a radio technician school to learn to build a radio receiver. It was here that Backus discovered his aptitude for mathematics and enrolled himself at Columbia University to study maths. There, the Selective Sequence Electronic Computer (SSEC) — a one-of-a-kind room-sized electromechanical computer — was on display. Backus made a passing mention of his interest in mathematics to his tour guide, unaware himself that he was sowing the seeds for his life’s work. An informal oral examination made up of mathematical brainteasers followed, and, owing to his excellent results, he was taken upstairs to meet the SSEC project director. A person with little-to-no idea of what programming entailed was instantly hired as a programmer. He would go on to become a pioneer in the field. “You need the willingness to fail all the time. You have to generate many ideas only to discover that they don’t work. And you keep doing that over and over until you find one that does work. John Backus Even though it hadn’t been evident so far in his life, Backus soon realised that he was an expert at solving problems. He not only had a knack for finding ways of breaking down seemingly difficult problems into a more manageable form, but also had a willingness to fail in his search for knowledge. During his first three years at IBM, there were a number of instances when this came to the fore. With the SSEC tasked with the calculation of ephemeris tables, Backus helped develop techniques for the same. These were later employed by NASA for the Apollo lunar missions of the 1960s. To facilitate the process of writing instructions at the machine level, which is what programming revolved around back then, Backus invented Speedcoding. This allowed operations on floating point numbers to be described in a symbolic form. Aware of how cumbersome programming was and based on his Speedcoding work, Backus proposed an idea. His diverse group included a chess wiz, a researcher from the Massachusetts Institute of Technology, a cryptographer, and a crystallographer. Not to be left out, the team also included Lois Haibt, fresh out of Vassar College and the only woman in the team. Right from the start, Backus knew that IBM’s method of having yearly performance reviews was not suited for his team. Not new to rule-breaking, he decided to completely ignore the whole process and offered his team the freedom they required. As a team, they were looking to address two fundamental programming problems. Firstly, they wanted to make programming accessible to a broader range of people by making it both cheaper and faster. Secondly, they had to figure out how to structure the underlying code to make this possible. “Much of my work has come from being lazy. I didn’t like writing programs, so I started work on a system to make them easier to write. John Backus Backus, however, had another way of summing this all up.
Less than a year later, however, he was expelled due to his poor attendance. In 1945, he continued to study medicine at Flower and Fifth Avenue Medical School in New York, only to drop out again nine months later “because all you had to do was memorise stuff. Having come through that somehow, Backus moved to the University of Virginia to study chemistry in 1942. He had another operation on his head to replace the plate with one he made himself and left the army in 1946 with an honourable medical discharge. Just before his graduation in 1949, Backus found himself wandering into IBM’s headquarters on New York’s Madison Avenue one day. IBM was working on IBM 704 at the time, a scientific computer primarily designed for floating point operations. He suggested creating a new language that would make it easier to program the 704. Cover of the first reference manual for Fortran. | Photo Credit: Wikimedia Commons Once it was accepted, Backus built a small team of 10 people that worked out of the IBM World Headquarters in Manhattan and sought to make programming better.
“You need the willingness to fail all the time,” Backus once said. “Much of my work has come from being lazy,” he later said.
